AI/ML Engineer

Role: AI/ML Engineer – LLM Architecture, RAG, and Agentic AI

Exp : 8 +yrs

Location : Remote

About Cognia Security

cutting-edge, AI-native cybersecurity platform

Our product — uses LLMs, context graphs, and real-time ingestion to surface material risks across code, infrastructure, identity, and data. We’re solving problems security engineers face daily — and giving every engineer an always-on expert assistant.

We’re looking for an experienced AI/ML Engineer (Architect level) who can design and implement the AI platform framework that powers Cognia’s intelligence layer. This includes architecting RAG pipelines, LLM memory systems, and Agentic AI flows.

If you thrive in zero-to-one environments, care deeply about product + engineering, and want to help shape the future of autonomous security — we want to hear from you.

Responsibilities

• Architect and build Cognia’s AI framework, integrating LLMs, memory, RAG, and agentic reasoning across the platform.
• Design AI agents and reasoning flows that interpret app, infra, and IAM telemetry for risk triage and remediation.
• Develop retrieval-augmented generation (RAG) pipelines for context-based summarization, persona-driven views, and continuous learning.
• Implement LLM memory and context management (short-term, long-term, and episodic memory) for persistent reasoning.
• Tune prompts, evaluation logic, and reasoning paths across multiple models (GPT-4, Claude, Gemini, etc.).
• Collaborate with founders, backend engineers, and data teams to integrate AI logic with product pipelines and APIs.
• Define AI architecture standards, evaluation metrics, and observability for all LLM-driven components.

Requirements

• 5–8 years of experience in AI/ML or backend systems, with a strong focus on LLM application development and AI system design.
• Proven experience architecting and deploying AI frameworks or agentic systems in production environments.
• Strong expertise in AI framework and the Java ecosystem — comfortable building and orchestrating AI workflows using Spring Boot, Spring AI, and vector storage integrations.
• Experience in RAG architecture, prompt engineering, memory management, and model orchestration patterns.
• Familiarity with OpenAI, Anthropic, and Google Gemini APIs and evaluation frameworks for reasoning quality.
• Strong software engineering fundamentals — versioning, pipelines, CI/CD, and API integration.

• Experience with vector databases (e.g., Pinecone, Weaviate, or Chroma) or knowledge graph-based retrieval.
• Bonus:
• Background in cybersecurity, SaaS platforms, or risk intelligence systems.